Abstract Effective predictive biomarkers are needed for checkpoint inhibitors (ICI) based-immunotherapy in non-small-cell lung cancer (NSCLC). This study aims to evaluate whether single nucleotide polymorphisms (SNPs) in programmed cell death-ligand 1 (PD-L1) might effectively predict tumor response to ICI in Caucasian patients with advanced NSCLC. Peripheral blood samples and clinico-pathological data were prospectively collected from advanced NSCLC patients treated with anti-Programmed cell death protein 1 (PD-1) inhibitor nivolumab. rs822336, rs2282055 and rs4143815 PD-L1 SNPs were analysed based on previous reports available in the literature. PD-L1 SNPs were analysed following DNA extraction from stored peripheral blood lymphocytes by Taqman Real-Time PCR assay. Presence of PD-L1 SNPs were correlated with clinico-pathological data. Mechanisms underlying the effect of PD-L1 SNPs on PD-L1 expression were analysed in vitro utilizing five NSCLC cell lines, RNA expression and flow cytometry analysis as well as PROMO bioinformatic tool. PD-L1 SNPs localization were analysed utilizing Ensembl genome browser. Differences were considered statistically significant with p value< 0.05. Twenty-eight patients were included in the study. rs2282055 was present as wild type (wt) TT (rs2282055>T), GT and GG genotype in 56.25%, 34.38% and 9.38%, respectively, of patients. On the other hand, rs822336 was present as wt GG (rs822336>G), GC and CC genotype in 28.13%, 37.50% and 34.38%, respectively, of patients. Progression free survival (PFS) was significantly correlated with presence of both PD-L1 SNPs rs2282055 and rs822336 but not with rs4143815. Patients carrying GG either for rs2282055 or rs822336 had a significantly (p=0.0434 and p=0.0131, respectively) shorter PFS as compared to those carrying other PD-L1 SNP genotypes. NSCLC cell lines carrying either GG for rs2282055 or rs822336 significantly (p<0.05) expressed a higher level of PD-L1 expression as compared to cells with other PD-L1 SNP genotypes. Interferon gamma treatment induced PD-L1 expression regardless of PD-L1 SNPs. rs822336 and rs2282055 mapped in enhancer-promoter and intronic regions of PD-L1 gene, respectively. rs822336 was further investigated because of its highest clinical significance and gene localization. Twelve transcription factors were identified to potentially bind rs822336. Two of them discriminated between alternative alleles at the SNP locus and are further under investigation. Previous studies have shown the role of SNPs in cancer patient outcome as well as in mRNA and protein expression changing. In the present study we demonstrated that some of PD-L1 SNPs can affect the basal expression of PD-L1. These results have clinical significance since PD-L1 SNPs predict tumor response to nivolumab-based immunotherapy in Caucasian patients with advanced NSCLC. Citation Format: Francesco Sabbatino, Giovanna Polcaro, Luigi Liguori, Valentina Manzo, Valeria Conti, Jessica Dal Col, Anita Vergatti, Cristiana Stellato, Vincenzo Casolaro, Amelia Filippelli, Soldano Ferrone, Stefano Pepe. Single nucleotide polymorphisms (SNPs) in PD-L1 as predictive biomarkers for checkpoint inhibitor based-immunotherapy in caucasian patients with advanced NSCLC [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 3132.